Endoscopic, Endonasal Variability in the Anatomy of

Peer-Review Reports

Endoscopic, Endonasal Variability in the Anatomy of the Internal Carotid Artery He´le`ne Cebula1,4,*, Almaz Kurbanov1,3,*, Lee A. Zimmer1-3, Pavel Poczos1,3, James L. Leach5, Juan Carlos De Battista6, Se´bastien Froelich7, Philip V. Theodosopoulos1,3,8, Jeffrey T. Keller1,3,8

Key words Anatomic study - Classification - Endoscopic endonasal approaches - ICA segments - Pituitary gland

- BACKGROUND:

Abbreviations and Acronyms 2D: Two dimensional 3D: Three dimensional ICA: Internal carotid artery

- METHODS:

-

From the Departments of 1 Neurosurgery and 2 Otolaryngology—Head and Neck Surgery, University of Cincinnati College of Medicine, and 3Brain Tumor Center, University of Cincinnati Neuroscience Institute, Cincinnati, Ohio, USA; 4Department of Neurosurgery, Hautepierre University Hospital, Strasbourg, France; 5Department of Radiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA; 6Department of Neurosurgery, Pelegrina Hospital and Ossys Institut, Sección (Mendoza), Argentina; 7Department of Neurosurgery, Lariboisiere University Hospital, Paris, France; and 8Mayfield Clinic, Cincinnati, Ohio, USA To whom correspondence should be addressed: Jeffrey T. Keller, Ph.D. [E-mail: [email protected]] Citation: World Neurosurg. (2014) 82, 6:e759-e764. http://dx.doi.org/10.1016/j.wneu.2014.09.021

Classic three-dimensional schemas of the internal carotid artery (ICA) for transcranial approaches do not necessarily apply to twodimensional endoscopic views. Modifying an existing ICA segment classification, we define endoscopic orientation for the lacerum (C3) to clinoid (C5) segments through an endonasal approach.

In 20 cadaveric heads, we classified endoscopic appearance based on shape and angulation of C3 to C5 segments. Distances were measured between both arteries, and between the ICA and pituitary gland.

- RESULTS:

We identified 4 common ICA patterns: types I through III matched side-to-side, whereas type IV was asymmetric. In 80% of specimens, the pituitary gland had direct contact with the ICA. In 20% of specimens, a space existed between the pituitary gland and the cavernous segment. Access to the posterior aspect of the cavernous sinus medial to the cavernous segment was possible without retraction of the artery or pituitary gland. Spaces between the lacerum and cavernous segments were trapezoid (80%) and hourglass (20%).

- CONCLUSIONS:

Distinguishing which ICA type courses between the lacerum and clinoid segments can help clarify the relationships between the artery and its surrounding structures during endoscopic approaches. Adapting the classic terminology of ICA segments provided consistency of endoscopic relevance, defined potential endoscopic corridors, and highlighted the critical step of arterial contact.

Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2014 Published by Elsevier Inc.

INTRODUCTION Classification schemas of the internal carotid artery (ICA) have been proposed largely for application in transcranial approaches based on three-dimensional (3D) views of the ICA and its segments. These same classifications do not necessarily apply to two-dimensional (2D) views in endoscopic approaches where orientation and identification of the exact course and varied angulations of these arterial segments becomes difficult. Endonasal endoscopic approaches allow surgical access to various areas of the skull base including the planum sphenoidale, sellar floor, cavernous sinus, clivus, and petrous apex. However, this complex region has not yet been well-classified endoscopically. Such

anatomic detail through the endoscopic view is critically important because injury to the ICA poses significant morbidity and mortality, reportedly ranging from 0e3.8% in endoscopic transsphenoidal procedures (6). Furthermore, the ICA is the first major structure exposed in the sphenoid sinus in endoscopic views, whereas its exposure follows that of the cranial nerves III, IV, and VI in transcranial views. In contrast with the classic 3D views of the ICA segments during open cranial surgery, 2D orientation in endonasal endoscopic views is often difficult, especially for identification of the often varied angulation and course of these segments. Commenting on the unpredictable nature of anatomy, Cappabianca (3) noted, “it changes with the perspective and direction of viewing a specific structure, the level of magnification and possibly anatomical variations of landmark reference points.” Given the predictability of such variants,

WORLD NEUROSURGERY 82 [6]: e759-e764, DECEMBER 2014

a comprehensive understanding of the ICA course and its relationship with the pituitary gland is essential to avoid injury. Several classification schemas have been proposed to identify various segments as well as the shape and/or course of the ICA afforded by a transcranial (13, 14, 22) or endoscopic endonasal view (10, 23) or angiographically (8, 18). Several classification/nomenclatures applied to endonasal endoscopic perspectives by Herzellah and Casiano (10) and Alfieri and Jho (1) were each anatomically correct, but lacked in agreement with each other and with intracranial ICA nomenclature. Adopting the Bouthillier et al. (2) classification, Fortes et al. (7) defined the anatomic landmarks, limitations, and difficulties of obtaining ICA exposure by endonasal endoscopic approaches, specifically in the distal parapharyngeal segments through a transpterygoid corridor.

www.WORLDNEUROSURGERY.org

e759

PEER-REVIEW REPORTS HÉLÈNE CEBULA ET AL.

ENDOSCOPIC INTERNAL CAROTID ARTERY

Proposing to modify an existing intracranial ICA nomenclature to an endonasal endoscopic perspective, we adapted the intracranial ICA classification system (C1-C7) proposed by Bouthillier et al. (2) and recently modified by De Powell et al. (5) for the endoscopic ICA as a template. Our cadaveric study examined anatomic variations of the specific angles and course of the ICA lacerum (C3) to the clinoid (C5) segments observed by endonasal endoscopic views. We identify the relationship between the left and right ICAs and the artery and pituitary gland, note the shapes of C4 and C4-C5 bends, and characterize the intercarotid space between the lacerum (C3) and cavernous (C4) segments. METHODS In 20 formalin-fixed cadaveric heads, specimens (40 sides) were injected with colored latex as previously described (21). Endoscopic transnasal dissections were performed using a rigid 4-mm-diameter endoscope, 14 cm in length with 0 and 30 rod lenses (Medtronic, Jacksonville, Florida, USA). The endoscope was connected to a light source through a fiber optic canal and a video camera, which was connected to a video monitor; images were captured using a high-definition digital video system (Stryker, Kalamazoo, Michigan, USA). In all specimens, the ICA was exposed bilaterally from the lacerum (C3) to the clinoid (C5) segments. We documented the shape and course of the ICA along the length of each specimen, and measured the distances between the left and right arteries and between the ICA and pituitary gland. We defined ICA segments C3-C5 from an endoscopic perspective based on De Powell et al.’s (5) system that describes 4 ICA bends called the C2, C3-C4, C4, and C4-C5 bends—a modification of Bouthillier et al. (2) original classification of ICA segments (Figure 1). We then used the subdivisions of the cavernous (C4) ICA, as described by Debrun et al. (4), into 5 parts: posterior ascending portion, junction of the posterior ascending and horizontal portions, horizontal portion, junction of the horizontal and anterior ascending portions, and anterior ascending portion.

e760

www.SCIENCEDIRECT.com

Figure 1. Conceptual illustration of the endoscopic perspective depicts the various internal carotid artery (ICA) classifications. (Left) Bouthillier et al. (2) used 7 segments: C1 ¼ cervical, C2 ¼ petrous, C3 ¼ lacerum, C4 ¼ cavernous, C5 ¼ clinoid, C6 ¼ ophthalmic, and C7 ¼ communicating. (Right) De Powell et al. (5) modification includes C3-C4 bend, C4 bend, and C4-C5 bend. Depending on the angle of the C4 bend (green plane), a potential corridor between the ICA and the pituitary allows access to the posterior cavernous sinus (yellow arrow). SOF, superior orbital fissure; OS, optic strut; OCR, opticocarotid recess; TS, tuberculum sellae. (Printed with permission Mayfield Clinic.)

Several characteristics were studied to classify the shapes of the ICA from an endoscopic 2D perspective. Specifically, the angle of the C4 bend was formed at the transition from the posterior ascending portion to the horizontal segment of the cavernous ICA (Figure 2). We then classified the angles into three types as 100 (type III) (Figure 2). Using a custom-designed caliper (Integra Life Sciences, Cincinnati, Ohio, USA) and a metric ruler, we measured the distances between the ICAs and between the pituitary gland and each ICA. Specifically, the distances between the left and right ICAs were measured at 5 anatomic levels: at the distal dural ring; the most concave point of the C4-C5 bend; at the most convex point of the C4 bend; at the C4 posterior ascending portion; and at the foramen lacerum (Figure 3A). Distances between the pituitary gland and ICAs were measured at 3 levels: at the distal dural ring, most concave point of the C4-C5 bend, and most convex point of the C4 bend (Figure 3B).

RESULTS Shapes of the ICA In 20 cadaveric specimens (40 sides), we defined 4 characteristic shapes: types I through IV extended from C3 to the C5 segments based on the angle between the posterior ascending and the horizontal portions of the C4 segment (Figure 4). Type I. Type I (25%) have the most prominent tortuosity at the C4 bend, where the posterior ascending portion of the ICA approaches the posterior clinoid process. The angle between the posterior ascending and the horizontal portions of the C4 segment was less than 80 (Figure 4A). The C4 and C4-C5 bends were more prominent and tortuous than the other types. Furthermore, the C4 bend was always in contact with the pituitary gland. In half of type I specimens, the C4 bend was posterior to the pituitary gland and could not be visualized using the endoscope. The horizontal portion of the ICA had an inferior and lateral course.

WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2014.09.021



posterior ascending and the horizontal portions exceeded 100 (Figure 4C) with the C4 bend readily visualized. The ICAs were in contact with the pituitary gland in 66% of the specimens, 33% along the horizontal portion of C4 segment, and 33% at the C4 bend. The horizontal portion of the C4 segment had a superior and lateral direction. Overall, the bilateral ICAs in type III were less tortuous than seen in either type I or II. Type IV. The asymmetric type IV (25%) differs, specifically between the angle of the posterior ascending and the horizontal portion of C4 of the left and right arteries (Figure 4D).

Figure 2. Endoscopic view of the internal carotid artery showing 3 types of angles (black lines) between the posterior ascending and horizontal portions of the C4 segment. (A) angle 100 ; and (C) angle between 80 and 100 . PG, pituitary gland; ON, optic nerve. *C4 bend. (Printed with permission from Mayfield Clinic.)

Type II. In type II (35%), the posterior ascending portion of the C4 segments on both sides was more vertical than in type I. The angle between the posterior ascending and the horizontal portions of the C4 segment was between 80 and 100 (Figure 4B). In 78% of specimens, the C4 segment was in contact with the pituitary gland: 21% at the C4 bend and 57% along its horizontal segment. In 14% of cases, the C4 bend projected posterior to the pituitary gland. Its horizontal portion had a lateral and slightly superior course. Type III. Type III (15%) ICAs appeared slightly curvilinear. The angle between the

Intercarotid Space The intercarotid spaces measured between the right and left ICAs were range (mean) from 1) 9e25 mm (14.95 mm) at the level of the distal dural ring; 2) 13.5e30 mm (19.53 mm) at the most concave point of C4-C5 bend; 3) 3e19 mm (12.15 mm) at the level of the most convex point of C4 bend; 4) 13e22 mm (17.35 mm) at the C4 posterior ascending portion; and 5) 11e26 mm (17.2 mm) at the foramen lacerum level (Figure 3A). Two different shapes, trapezoid and hourglass, were observed for the area between the ICAs at the C3 and C4 levels. For type IeIII arteries, the shape was trapezoid in 80% of cases and hourglass in 20% (Figure 5). The course of the posterior ascending portion of C4 varied between the two shapes: the course was superior and medial as it approached the posterior clinoid process in trapezoid-shaped specimens but was superior and lateral in hourglass-shaped specimens. Relationship Between the Pituitary Gland and the ICA Overall, the C4 segment was in contact with the pituitary gland in 40% of specimens along the C4 bend, in 7.5% along the horizontal portion, in 32.5% where both the horizontal and the C4 bend contacted with the pituitary, and absent in 20% of specimens (Figure 6). The distance between the pituitary gland and ICA was measured at 3 levels on each side. Average distances at the level of the distal dural ring, at the most concave point of the C4-C5 bend, and at the level of the most convex point of the C4 bend were 0.6,


3.63, and 0.65 mm, respectively, on the right and 0.7, 3.98, and 0.65 mm, respectively, on the left. DISCUSSION Our endoscopic classification of the ICA defined 4 characteristic arterial shapes as types IeIV that extended from C3 to C5 segments based on the angle between the posterior ascending and the horizontal portions of the C4 segment. Our adaptation of an existing ICA classification for transcranial approaches (later modified by De Powell et al. [5]) proved effective in characterizing segments visualized from an endoscopic 2D perspective (Figure 1). Careful examination of the course and shape of the ICA bilaterally in the cadaveric specimens enabled characterization of the anatomic variability of the ICA course. Specifically, the well-known 3D landmarks of the sphenoid sinus, including the lateral opticocarotid recess, medial opticocarotid recess, optic prominence, carotid prominence, clival recess, and prominence of the sellar floor are now better defined for 2D endoscopic views relative to the angulation and trajectory of the ICA. Shapes of ICA The 4 characteristic ICA shapes (symmetric in types I through III and asymmetric in type IV) can be identified preoperatively using magnetic resonance angiography/ computed tomographic angiography. During surgical planning, this classification can help in anticipation of potential vascular injury. We speculate that type I presents the highest risk for vascular injury based on its contact between the ICA and pituitary gland. In 50% of our specimens, the C4 bend was behind the pituitary gland (Figure 4A). Risk of potential vascular injury decreases in types II and III. This is especially true for the latter, which ensures the surgeon of an existing space between the ICA and the pituitary gland. Several investigators described the course of the ICA from a transcranial lateral view (13, 17, 22). The 3 shapes described by Lang (17)—an S-shape in 30%, straight in 17%, or an intermediate form in 53% in transcranial approaches—were similar to our ICA types I through III by endoscopic views. Describing 2 curvatures


e761



Figure 3. Anatomic measurements between the internal carotid arteries and the pituitary gland in 20 specimens. (A) (aee) Intercarotid distances between the left and right ICAs. (B) Measurements (a’, b’, c’) of the space between the ICA and the pituitary gland at 3 levels (cephalic, middle, caudal). (Printed with permission from Mayfield Clinic.)

of the ICA through a transcranial approach, Jittapiromsak et al. (13) observed either a normal-type or a redundant-type curvature (a tortuous configuration), and found “the pattern of curvature was similar bilaterally.” In several endonasal endoscopic descriptions of the ICA, Herzallah and Casiano (10) described 2 shapes of ICAs based on the intracavernous artery. Type 1, a classic shape occurred most often (70%) with the C4 bend described as “a simple transition between the adjacent segments,” whereas type 2 (30%) was a C-like shape at the C4-C5 bend that was comparable to our type I (6). They subdivided their type 1 into subtypes a and b based on the angle of the C4 bend; specifically, type 1a (right angle at C4 bend) and type 1b (an obtuse angle at the C4 bend) corresponds to our types II and III, respectively. They observed that both ICAs were symmetric in shape. Yilmazlar et al. (23) described a “medial curled course of ICA” in 28.6% of cadavers. This shape corresponded to our type I in both shape and frequency of

e762


occurrence. Fortes et al. (7) described various surgical corridors related to the ICA segments C1-C4, specifically comparing the shapes of the ICA bilaterally and noting tortuosity. In an angiographic analysis to characterize ICA shape, Lazorthes (18) distinguished 4 types as significant tortuosity (49.1%), less tortuous (24.1%), almost straight (23.6%), and straight (3.2%). The two tortuous types were similar to our types I and II, whereas their nearly straight type corresponded to our type III. In another angiographic classification, Francke et al. (8) described 3 types as fetal, simple, and double siphon, and observed that ICA tortuosity increases with age. Recent studies (10, 13) have reported that the ICAs are symmetric bilaterally. However, observing that the ICAs were not always mirror images of one another in our specimens, we designated type IV to describe this asymmetry. Knowledge regarding ICA shape is essential for surgical considerations (i.e., anticipate these variations during a transsphenoidal endoscopic approach).

Relationship Between the Pituitary Gland and the ICA The space between the ICA and the pituitary gland varies depending on the anatomy of both structures (14). We focused on this space because tumors in this region can invade surrounding structures (e.g., pituitary adenomas can grow or invade the medial wall of the cavernous sinus). The medial wall may consist of tough, tenacious connective tissue or may be fenestrated, incomplete, or essentially nonexistent, offering little or no anatomic resistance against tumor invasion (15). Surgical access to the medial and posterior wall of the cavernous sinus is possible by elevating or retracting the pituitary gland medially or by retracting the C4 or C4-C5 bend of the ICA laterally. In 20% of cadaveric specimens, there was a space between the pituitary gland and the cavernous ICA that facilitated access to the posterior aspect of the cavernous sinus (Figure 6). This space serves as a corridor to access the posterior aspect of the cavernous sinus, to remove tumor that




Figure 6. Transnasal transsphenoidal endoscopic view of a type II angle (between 80 and 100 ) that has no contact with the pituitary gland. Angle allows a corridor to the posterior aspect of the cavernous sinus and the oculomotor nerve without retraction of the internal carotid artery or the pituitary gland. CN III, oculomotor nerve; CS, cavernous sinus; PG, pituitary gland. (Printed with permission from Mayfield Clinic.)

Figure 4. Transnasal transsphenoidal endoscopic view of the parasellar region illustrate that types IeIII are symmetric and type IV is asymmetric. (A) Type I angle between the posterior ascending and horizontal portions of C4 segment is 100 . ICA appears slightly curvilinear and less tortuous than the type I or the type II. (D) Type IV angles of the left and right ICAs are asymmetric. PG, pituitary gland; ON, optic nerve. *C4 bend. (Printed with permission from Mayfield Clinic.)

extends into the posterior cavernous sinus, or to represent a potentially safe surgical corridor lateral to the pituitary gland. Other investigators have described triangles that serve as corridors to the cavernous sinus and the paracavernous region (11, 12). We suggest that the corridor to the posterior and lateral

cavernous sinus can be called the medial triangle of the cavernous sinus. Appreciation of this anatomic window is important when access to the cavernous sinus is necessary for tumor removal, especially because this region is difficult to reach using a transcranial approach.

Figure 5. Transnasal transsphenoidal endoscopic view between the C3 and the C4 segments of the internal carotid artery at the lacerum and clivus levels. Two distinct shapes (green) were identified as trapezoid (A) in 80% or hourglass (B) in 20% of specimens. (Printed with permission from Mayfield Clinic.)


Distance Between C3 and C4 Segments The endoscopic transsphenoidal, transclival approach described by Fraser et al. (9) allows access to pathologies of the ventral brainstem (i.e., chordomas, chondrosarcomas without significant lateral extension). It serves as the most direct route to ventral pathologies of the posterior fossa. Its disadvantage is its lateral limitation owing to the position of the ICAs and the risk of skeletonizing them. In other studies (16, 19, 24) of the endoscopic transsphenoidal, transclival approach to reach pathology in the petrous apex, investigators have stressed the importance of understanding the relationship of the ICA and the tumor. Tumors that extend posterior to the lacerum segment are difficult to remove because the ICA is fixed/bound to the underlying fibrocartilaginous contents of the foramen lacerum. Extensive drilling of clivus and petrous bone is necessary to access the space behind the C3 segment. In contrast, the C4, horizontal segment, and C4 bend were easily skeletonized and retracted laterally in cadavers allowing posterior access to the medial posterior cavernous sinus. In 20% of our specimens, we observed a space between the inferior aspect of the pituitary gland and the C4 segment that could allow access to and removal of the posterior clinoid process and upper clivus. Renn and Rhoton (20) report that 82% of cases had the shortest distance between both ICAs in the supraclinoid area, which is very similar to our findings. The space


e763


between the left and right C3 and C4 segments was trapezoidal in 80% of specimens and hourglass in 20% (Figure 5). With knowledge of the possible variations of the ICA, the surgeon may anticipate the artery’s course and possible variations, thus planning for safe drilling of the clivus and avoiding vascular injury. Clinical Relevance Increasing use of endoscopic transnasal approaches has renewed a sharp focus on the anatomic variations of the ICA’s course along the skull base. Endoscopic approaches to the pituitary region are now common. We and other investigators have also promoted it as a safe, effective approach for a variety of anterior and middle skull base lesions, centrally including all clival lesions and laterally along the pterygopalatine and infratemporal fossas, orbital apex, petrous apex, and cavernous sinus. Corridors medial and lateral to the ICA have been described and studies have reported on lesions anterior and posterior to the ICA in any part of the skull base. Most important to ensuring the safety of these endoscopic surgical approaches during this explosion of interest is a fundamental, practical understanding of the ICA anatomy, normal and its variations, as viewed from the anterior transnasal approach. Because many lesions along the anterior and middle skull base do distort the normal ICA anatomy and, on occasion, engulf and narrow it, this anatomic understanding becomes a necessary tool in maximizing extent of resection and safety of surgical approach. Careful study of preoperative imaging including magnetic resonance imaging, computed tomographic scans with thin cuts through the skull base, computed tomographic angiograms, magnetic resonance angiograms, and on occasion digital subtraction 2D and 3D angiograms is imperative in the preoperative evaluation of any skull base lesion to be approached in an endoscopic manner. CONCLUSIONS Using an endonasal endoscopic approach, we classified the ICA segments from the C3 to the C5 as types I through IV by applying an existing ICA nomenclature schema defined by Bouthillier et al. (2) and later modified by

e764



De Powell et al. (5). Our 4 ICA types are relevant for endoscopic views, specifically the relationship between the ICA and the pituitary gland, as well as the artery’s course from the C3 to the C5 segments. Understanding the anatomic variability of the ICA allows for better preoperative planning and decreases the risk of vascular injury during endoscopic approaches to the skull base during an endonasal approach. REFERENCES 1. Alfieri A, Jho HD: Endoscopic endonasal cavernous sinus surgery: an anatomic study. Neurosurgery 48:827-837, 2001. 2. Bouthillier A, van Loveren HR, Keller JT: Segments of the internal carotid artery: a new classification. Neurosurgery 38:425-433, 1996. 3. Cappabianca P: The lesson of anatomy. Surg Neurol 71:597-599, 2009. 4. Debrun G, Lacour P, Vinuela F, Fox A, Drake CG, Caron JP: Treatment of 54 traumatic carotidcavernous fistulas. J Neurosurg 55:678-692, 1981. 5. De Powell J, Froelich S, Zimmer LA, Leach JL, Karkas A, Theodosopoulos P, Keller JT: Endoscopic transnasal vs. open transcranial internal carotid artery anatomy: a uniform application of nomenclature. World Neurosurg. Published online 17 September 2014. 6. Dusick JR, Esposito F, Malkasian D, Kelly DF: Avoidance of carotid artery injuries in transsphenoidal surgery with the Doppler probe and micro-hook blades. Neurosurgery 60 (4 Suppl 2): 322-329, 2007. 7. Fortes FS, Pinheiro-Neto CD, Carrau RL, Brito RV, Prevedello DM, Sennes LU: Endonasal endoscopic exposure of the internal carotid artery: an anatomical study. Laryngoscope 122:445-451, 2012. 8. Francke JP, Macke A, Clarisse J, Libersaet JC, Dobbelaere P: The internal carotid arteries. Surg Radiologic Anat 3:243-261, 1982.

carotid artery to intracavernous neural structures. Skull Base 20:327-336, 2010. 14. Kitan M, Taneda M, Shimono T, Nakao Y: Extended transsphenoidal approach for surgical management of pituitary adenomas invading the cavernous sinus. J Neurosurg 108:26-36, 2008. 15. Knappe UJ, Konerding MA, Schoenmayr R: Medial wall of the cavernous sinus: microanatomical diaphanoscopic and episcopic investigation. Acta Neurochir (Wien) 151:961-967, 2009. 16. Komotar RJ, Starke RM, Raper DM, Anand VK, Schwartz TH: The endoscope-assisted ventral approach compared with open microscope-assisted surgery for clival chordomas. World Neurosurg 76:318-327, 2011. 17. Lang J: Transnasal approach to the pituitary region. In: Lang J, ed. Clinical Anatomy of the Head: Neurocranium, Orbit, Craniocervical Regions. New York: Springer-Verlag; 1983:146-170. 18. Lazorthes G: Vascularisation et circulation cérébrale [in French]., Première partie: La vascularization cérébrale; Chapitre premier: Les artères carotide interne et vertébrale. Paris: Masson & Cie; 1961:7-24. 19. Paluzzi A, Gardner P, Fernandez-Miranda JC, Pinheiro-Neto CD, Scopel TF, Koutourousiou M, Snyderman CH: Endoscopic endonasal approach to cholesterol granulomas of the petrous apex: a series of 17 patients: clinical article. J Neurosurg 116:792-798, 2012. 20. Renn WH, Rhoton AL Jr: Microsurgical anatomy of the sellar region. J Neurosurg 43:288-298, 1975. 21. Sanan A, Abdel Aziz KM, Janjua RM, van Loveren HR, Keller JT: Colored silicone injection for use in neurosurgical dissections: anatomic technical note. Neurosurgery 45:1267-1274, 1999. 22. Willis T: The Anatomy of the Brain. 1681. Reprinted by USV Pharmaceutical Corp. Tuckahoe, New York: USV Pharmaceutical Corp; 1971. 23. Yilmazlar S, Kocaeli H, Eyigor O, Hakyemez B, Korfali E: Clinical importance of the basal cavernous sinuses and cavernous carotid arteries relative to the pituitary gland and macroadenomas: quantitative analysis of the complete anatomy. Surg Neurol 70:165-175, 2008.

9. Fraser JF, Nyquist GG, Moore N, Anand VK, Schwartz TH: Endoscopic endonasal minimal access approach to the clivus: case series and technical nuances. Neurosurgery 67 (3 Suppl Operative):ONS150-ONS158, 2010.

24. Zanation AM, Snyderman CH, Carrau RL, Gardner PA, Prevedello DM, Kassam AB: Endoscopic endonasal surgery for petrous apex lesions. Laryngoscope 119:19-25, 2009.

10. Herzallah IR, Casiano RR: Endoscopic endonasal study of the internal carotid artery course and variations. Am J Rhinol 21:262-270, 2007.

Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

11. Inoue T, Rhoton AL Jr, Theele D, Barry ME: Surgical approaches to the cavernous sinus: a microsurgical study. Neurosurgery 26:903-932, 1990.

*both authors contributed equally.

12. Isolan GR, Karayenbuhl N, de Oliveira E, AlMefty O: Microsurgical anatomy of the cavernous sinus: measurements of the triangles in and around it. Skull Base 17:357-367, 2007.

Citation: World Neurosurg. (2014) 82, 6:e759-e764. http://dx.doi.org/10.1016/j.wneu.2014.09.021

13. Jittapiromsak P, Sabuncuoglu H, Deshmukh P, McDougall CG, Spetzler RF, Preul MC: Anatomical relationships of intracavernous internal

Received 25 February 2014; accepted 12 September 2014; published online 17 September 2014

Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2014 Published by Elsevier Inc.
